This application claims the priority benefit of European application serial no. 02009151.8, filed Apr. 24, 2002.
1. Field of the Invention
The present invention relates to a method of testing for the recognition of irregularities in the wall thickness of ferromagnetic pipes as well as the test apparatus especially designed for the implementation of the test method.
2. Description of the Related Art
In the oil and gas industry, mainly steel pipes are used to transport the product. Normally these are made from fern tic steel, which show distinct ferromagnetic characteristics. Corrosion, or wall thickness reduction, in the pipes is caused by the influence from external as well as internal factors. This may lead to product loss and, as a result, to environmental damage. As a preventive measure, the pipes are periodically tested so that wall thickness changes can be detected in time. Nowadays, the following methods are used to test these pipes: (i) the so-called intelligent pig and (ii) the so-called NoPig Method. Intelligent Pigs are, for example, described in the xe2x80x9cNon-destructive Testing Handbook, 2nd Ed., Vol. 10, Non-destructive Testing Overview, American Society for Non-destructive Testing, 1996, S. 252xe2x80x9d. An intelligent pig can be equipped with various sensors, utilizing ultrasound or magnetic flux leakage, to measure the actual wall thickness along the pipeline. The pig is inserted into the pipeline to be inspected and is transported through the line by the movement of the product flowing through the pipe. The measurement data are stored in the memory banks within the pig and will be evaluated once the pig has been removed from the line.
The disadvantage of the intelligent pig is the strict standards in which the pipelines must be built to accommodate the pig: for example, special channels must be available to insert and retrieve the pig, sharp bends, dents, different pipe diameters used in one pipeline are not allowed etc. The areas in which the pig can be efficiently used are therefore significantly reduced to the pipelines that have been specifically designed for this inspection method.
The U.S. Pat. No. 4,048,558 refers to a method in which a current will be passed through a metal pipe at different frequencies and its impedance is monitored. Should a change in the wall thickness occur, then the impedance should change as well. The disadvantage of this method is the minimal sensitivity and the problem that the defects can not be precisely located.
The German Patent Application DE 19819066 A1 refers to a non-contact inspection method for the recognition of wall thickness irregularities in inaccessible metal pipes. This method uses electric currents of different frequencies that flow through the pipe wall and induce magnetic fields that can be measured outside of the pipe. Through varying magnetic field penetration depths in the pipeline at lower and higher frequencies, the magnetic fields at these frequencies outside of the pipe will also vary when a defect is present of the pipe. This is caused by the deviation of the cross-sectional form of the pipe from the ring at the defect location. The current distribution in the cross section of the pipe will differ by varying frequencies so that the current weight center does not necessarily lie on the pipe symmetry axis. This causes the corresponding changes to the magnetic fields outside of the pipe. Above a pipe section containing no defects, the magnetic fields outside the pipe generated by varying frequencies will remain equal because the current centers will remain on the symmetrical axis of the pipe even by different frequencies. Defects will be found by comparing the measured magnetic field data at varying frequencies. It is necessary to sequentially scan the magnetic field along the pipe. This method is known as the NoPig Method (see http://www.finoag.com). This method does not have the disadvantage that pigs do in that no special preparation of the pipe is necessary and does not have to have the extra facilities to allow for the inserting of the inspection tool.
Nevertheless, the NoPig Method does have shortcomings. In the case of steel pipes, which tend to be made from ferritic steel, it is not satisfactory to measure the magnetic fields induced by two currents (one of them at a low frequency, and the other one with a higher frequency) and then to compare. Due to a rather high magnetic permeability of steel, the magnetic field in the pipe wall is much stronger than outside the pipe. In the case of a defect a stray magnetic flux arises from the defect area. As this takes place, the following describes the magnetic field outside the pipe: the stray magnetic flux acts against the displacement of the weight center of the current distribution. In the extreme case the magnetic field values to be measured are identical at low and high frequencies despite of an available defect. This compensating interaction of both mechanisms mentioned above depends on causes such as magnetic permeability and electrical conductivity of the steel used, and the geometric shape of a defect. This results in an essential reduction of the method and even in a full non-sensitivity to some defects.
In the light of the foregoing, it is an object of the present invention to provide a method and an apparatus for the recognition of irregularities in the wall thickness of ferromagnetic pipes, which makes it possible to, non-destructively and non-contacting, detect and to define places with reduced wall thickness on a pipe, despite of the fact that the pipe is made of a ferritic steel.
In accordance to the above objects and other advantages of the present invention, a non-destructive test method and a corresponding test apparatus for non-contacting recognition of irregularities in the wall thickness of ferromagnetic pipes are provided. Such irregularities are caused mainly by corrosion. For this reason the invention is especially intended for preventive recognition of corrosion caused wall thickness losses in oil and gas pipelines because they are made mainly from ferritic steel. The test method implies that an electric current consisting of many harmonic components is passed through the pipe tinder test and the magnetic field produced by the current will be measured. Therewith the frequency spectrum of the current is measured. The frequency spectrum of the measured magnetic field values is also calculated. The ratio between identical spectral components of the current and that of the magnetic field is calculated and used for the defect evaluation. The frequency range of the harmonic components is chosen in such a way, that the lowest frequency corresponds to the skin effect depth, which is at least as large as the nominal wall thickness of the pipe under test. The highest frequency corresponds to the skin effect depth which is at least as small as the critical residual wall thickness of the pipe under test. The critical residual wall thickness means the minimum allowed residual wall thickness on corroded areas. Since the ratio of the magnetic field outside the pipe at the place of corrosion to the current is frequency-dependent, this frequency dependence will be found from a comparison of spectral values of the magnetic field and of the current. In this way a defect will be recognized. In the case of a pipe section without any defects the ratio of the magnetic field outside the pipe to the current is frequency-independent. The test apparatus consists of two parts: a stationary current source and a mobile unit intended for magnetic field measurements. Both elements are transportable.